PNNL-27141
Review of Residential Comfort Control Products and Opportunities
December 2017
CE Metzger S Goyal MC Baechler
PNNL-27141
Review of Residential Comfort Control Products and Opportunities
CE Metzger S Goyal MC Baechler
December 2017
Prepared for the U.S. Department of Energy under Contract DE-AC05-76RL01830
Pacific Northwest National Laboratory Richland, Washington 99352
Executive Summary
Buildings consume about 40% of total energy used in the United States, and residential buildings account for more than half of the energy consumed in buildings.
Researchers have shown that about 60% of heating, ventilation, and air-conditioning (HVAC) systems have installation, commissioning, and performance problems, leading to as much as a 30% increase in annual energy consumption (Lstiburek 2010; Domanski 2014). The efficient design and improved operation of HVAC systems in homes could lead to significant energy reduction.
The development of commercially available sensors and control-related software in the past decade has resulted in a rapidly growing and changing HVAC control industry. Several papers describe parts of the HVAC controls space (Maasoumy, 2016, DOE 2016b, DOE 2016d, Wang and Goins, 2015). However, the objective of this paper is to fill a remaining need to help researchers capture a snapshot of HVAC control product capabilities and remaining opportunities that save homeowners and technicians’ time, money, and energy. By developing a taxonomy around the interaction between HVAC sensor and control technology, the current state-of-the-art product capabilities can be organized and summarized.
This paper begins by discussing the interaction of each major component in advanced sensor and control applications related to HVAC equipment. Figure ES.1 shows how each of these components interacts with each other to form appropriate inputs and outputs for a given application.
Sensors
Data Storage Computation Platform Human- and in-the- External Control Applications loop Signals
Actuators
Figure ES.1. Interactions between Several Component Technologies in an HVAC Control System (blue lines represent communication.)
The paper also looks at the applications of these components to commissioning, maintenance, and operations of the HVAC equipment in residential buildings. This paper assumes that commissioning is
iii
conducted by the same contractor who installed the HVAC system (or possibly by a third-party verifier). Scheduled maintenance, or retro-commissioning, are also included in this section. An overview of major commissioning applications—including the current state-of-the-art, the benefit of improving that current technology, and the possible missed opportunities associated with each application—is shown in Figure ES.2.
Figure ES.2. An Overview of State-of-art, Technical Gaps, and Benefits of Commissioning Applications
Maintenance is likely performed by an HVAC contractor. This is usually initiated by the homeowner, although some homeowners opt for a service contract providing biannual inspections and tune ups. A summary of the major maintenance applications is shown in Figure ES.3.
iv
Figure ES.3. An Overview of State-of-art, Technical Gaps, and Benefits of Maintenance Applications
v
HVAC operation revolves around the interaction of occupants and HVAC systems. Improving this interaction is a major objective of emerging intelligent controls. An overview of operations and optimization applications is shown in Figure ES.4.
Figure ES.4. An Overview of State-of-art, Technical Gaps, and Benefits of Operation Applications
A summary of state-of-the-art product features is also provided. These products are categorized through their primary application type (commissioning/maintenance or operation) and the features are categorized by component type (sensors, data storage, human-in-the-loop, communication, and controls).
A common theme that emerges from this study is the importance of connectivity between product categories. There are many manufacturers of sensors and many manufacturers of controls, but the power to save time and money through automation of any application comes with the ability to connect those manufacturers together through a standardized application programming interface (API).
vi
Two reoccurring opportunities to help save energy also appear throughout the paper. First, the use and control of variable speed motors throughout many components of the HVAC system would help take advantage of opportunities where full power is not necessary. Second, permanent and internet-connected sensors could help quickly diagnose and alert homeowners and technicians of non-optimal operating conditions so they can be remedied (e.g., dirty filters).
vii
Acknowledgments
This paper would not have been possible without funding from the U.S. Department of Energy’s Building America Program. The authors acknowledge and thank Eric Werling and Lena Burkett from the U.S. Department of Energy for their enthusiastic management of the Building America Program as well as their invaluable dedication to this paper. The authors also gratefully acknowledge help from Rosemarie Bartlett, Michael Brambley, Joshua Butzbaugh, Paul Ehrlich, Cory Fox, and Nora Wang for providing relevant information and excellent reviews.
ix
Acronyms and Abbreviations
API application programming interface ASHRAE American Society of Heating, Refrigerating and Air-Conditioning Engineers BTU British thermal unit(s) CO carbon monoxide
CO2 carbon dioxide EPA U.S. Environmental Protection Agency ERV energy recovery ventilation FDD fault detection and diagnostics GPS Global Positioning System HRV heat recovery ventilation HSRS HVAC Smart Relay Switch HVAC heating, ventilation, and air-conditioning IAQ indoor air quality IFTTT If This Then That kWh kilowatt-hour(s) NA not applicable PIR passive infrared PM particulate matter RECS Residential Energy Consumption Survey RH relative humidity UV ultraviolet VFD variable frequency drive VOC volatile organic compounds
xi
Contents
Executive Summary ...... iii Acknowledgments ...... ix Acronyms and Abbreviations ...... xi 1.0 Introduction ...... 1 2.0 State-of-the-Art Component Technologies ...... 3 2.1 Sensors ...... 3 2.2 Data Storage and External Signals ...... 6 2.3 Human-in-the-Loop ...... 6 2.4 Communication between Components ...... 7 2.5 Computation, Controllers and Actuators ...... 7 2.5.1 Smart/Intelligent Controllers and Actuators ...... 8 2.6 Integrated Sensors and Controls ...... 8 3.0 State-of-the-Art and Advanced Applications ...... 9 3.1 Commissioning ...... 9 3.1.1 Verifying Basic Functionality ...... 10 3.1.2 Duct Leakage and Air Flow ...... 11 3.1.3 Refrigerant Charge ...... 12 3.2 Maintenance ...... 13 3.2.1 Verifying Basic Functionality ...... 14 3.2.2 Refrigerant Level ...... 15 3.2.3 Air Filter Replacement ...... 15 3.2.4 Less Common Faults ...... 16 3.3 Improved Operation ...... 17 3.3.1 Operation Applications Focused on Comfort ...... 18 3.3.2 Operation Applications Focused on Energy Savings ...... 19 3.3.3 Operation Applications Focused on Grid Services Integration ...... 22 4.0 Summary of State-of-the-Art Product Features ...... 23 5.0 Remaining Technical and Market Opportunities ...... 31 5.1 Sensors ...... 31 5.2 Data Storage and External Signals ...... 31 5.3 Human-in-the-Loop ...... 32 5.4 Communication between Components ...... 32 5.4.1 Opportunities for Refrigerant Sensor Communication ...... 33 5.4.2 Automated System Zoning ...... 34 5.4.3 Opportunities for Ductless Mini-split Communication with Central Systems ...... 34 5.5 Computation, Controllers and Actuators ...... 34
xiii
5.6 Market Barriers ...... 35 6.0 References ...... 37
xiv
Figures
ES.1 Interactions between Several Component Technologies in an HVAC Control System ...... iii ES.2 An Overview of State-of-art, Technical Gaps, and Benefits of Commissioning Applications ...... iv ES.3 An Overview of State-of-art, Technical Gaps, and Benefits of Maintenance Applications ...... v ES.4 An Overview of State-of-art, Technical Gaps, and Benefits of Operation Applications ...... vi 1 U.S. Energy Consumption by Sector and End Usage ...... 1 2 Outline of Paper ...... 2 3 Interactions between Several Components in an HVAC Control System ...... 3 4 An Overview of State-of-art, Technical Gaps, and Benefits of Commissioning Applications ...... 10 5 An Overview of State-of-art, Technical Gaps, and Benefits of Maintenance Applications ...... 14 6 Summary of Improved Operation Applications ...... 17
Tables
1 Advantages, Disadvantages, and Applications of Several Types of Sensors ...... 4 2 Actuation Parameters Used for Controlling the HVAC Systems ...... 7 3 Remaining Opportunities to Reduce Time, Cost and Human Error Associated with Verifying the Basic Functionality of HVAC Systems ...... 11 4 Remaining Opportunities to Reduce Time and Cost Associated with Measuring Duct Leakage and Air Flow ...... 12 5 Remaining Opportunities to Reduce Time, Cost and Human Error Associated with Charging Refrigerant...... 13 6 Remaining Opportunities to Reduce Time and Cost Associated with Checking and Changing the Refrigerant Level ...... 15 7 Remaining Opportunities to Reduce Time and Cost Associated with Air Flow Issues in HVAC Systems ...... 16 8 Missed Opportunities to Improve IAQ while Minimizing Energy Use ...... 18 9 Missed Opportunities to Save Energy Associated with (De)Humidification ...... 19 10 Missed Opportunities to Save Energy with Natural Heating and Cooling ...... 20 11 Remaining Opportunities to Save Time and Cost When Installing Multi-Zone Systems ...... 21 12 Remaining Opportunities for Complete HVAC Optimization for Energy Reduction, Health, and Comfort ...... 21 13 Examples of Commissioning and Maintenance Refrigerant-Related Product Capabilities ...... 23 14 Examples of Air-Side Commissioning and Maintenance Product Capabilities ...... 25 15 Examples of Advanced Thermostat (Operation) Product Capabilities ...... 26 16 Examples of Indoor Air Quality (Operation) Product Capabilities ...... 28 17 Examples of Other Advanced Operation Products and Product Capabilities ...... 29
xv
1.0 Introduction
Buildings are one of the major energy consumers across the world. In the United States, they contribute approximately 40% to the total energy consumption (EIA 2015). According to the Residential Energy Consumption Survey (EIA 2009), residential buildings account for more than half of the energy consumed in buildings.
Figure 1 shows the total energy consumption in the United States by end-use sector and the energy consumption in homes by their end usage. As shown in the figure, heating and cooling consume about 48% of the energy usage in residential buildings. A small percentage of energy savings in the residential heating, ventilation, and air-conditioning (HVAC) systems can have high economic and environmental impacts.
a) Energy Consumption by End-use Sector b) Energy Consumption in Residential Buildings by End Usage
Figure 1. U.S. Energy Consumption by Sector and End Usage (EIA 2015 and EIA 2009)
To date, the majority of HVAC system sales have been replacement units that moved through a single distribution channel (EERE 2011). The manufacturer designs and builds the HVAC equipment, sells it to a distributor, who sells it to an HVAC contractor, who installs the unit for the end consumer. HVAC controls, such as thermostats and programmable thermostats, have typically been installed concurrently with the rest of the HVAC system. Now, more and more of these control devices are available at retail stores. This gives consumers more choices and pushes manufacturers to provide features that are valuable (e.g., saving time, money, or energy) for consumers.
Researchers have shown that about 60% of HVAC systems have installation, commissioning, and performance problems, leading to as much as a 30% increase in annual energy consumption (Lstiburek 2010; Domanski 2014). The efficient design and improved operation of HVAC systems in homes could lead to significant energy reduction. (As of 2015, there were approximately135 million households in the United States (EIA 2015), and the average annual electricity consumption for a U.S. residential customer is about 12,000 kWh. If 60% of homes in the U.S. were wasting 30% of their HVAC energy on performance issues that could be fixed with fault detection and diagnostics (FDD), about 146 billon kWh or 0.5 quads could be saved overall with better HVAC controls.)
While seven manufacturers (UTC/Carrier, Goodman/Amana, Trane/American Standard, Lennox, Rheem, York, and Nordyne) dominate the HVAC equipment market, they are not the market leaders in external HVAC controls. This disconnect between the equipment and controls manufacturers has left the market disjointed and flooded with options for consumers.
1
The objective of this paper is to capture a snapshot in time for this fast-moving industry. This paper will identify the current state-of-the-art sensing and control technologies, applications, and products. With this information and a review of recent literature, remaining opportunities to save homeowners and technicians time, cost, and energy are summarized. Although this paper touches on connected buildings and the “Internet of Things,” it does not provide a comprehensive review of those industries as some other review papers do (Maasoumy, 2016, DOE 2016b, DOE 2016d, Wang and Goins, 2015). This paper is different from other review papers because it focuses solely on potential advancements in residential HVAC control applications.
The organization of the paper is shown in Figure 2. It shows the topic for each section and indicates what question each section is attempting to answer. Section 2.0 reviews the existing state-of-the-art sensors and controls/actuators, communication strategies, computation platforms, and interfaces related to residential HVAC systems. A description of advanced sensing and control applications is presented in Section 3.0. Section 4.0 presents a national energy savings impacts assessment and market barriers evaluation of the applications and associated technologies. Section 5.0 quantifies the potential impact of each technology advancement on the overall energy use of U.S. housing stock and prioritizes gaps based on these estimates.
Section 2: State-of-the-Art Component Technologies What are the basic components of sensor and control technologies?
Section 3: State-of-the-Art and Advanced Applications How can the application of sensor and control technology be improved?
Section 4: Summary of Product Capabilities What product features are available today?
Section 5: Missed Opportunities and Market Barriers What are some opportunities beyond the current available products?
Figure 2. Outline of Paper
2
2.0 State-of-the-Art Component Technologies
A successful control system for implementing a basic or advanced control application may include up to six major components: (1) sensors, (2) data storage and external signals, (3) human-in-the-loop, (4) communication between components, (5) computation platform with control applications, and (6) actuators. Figure 3 shows interactions between these components.
Sensors
Data Storage Computation Platform Human- and in-the- External Control Applications loop Signals
Actuators
Figure 3. Interactions between Several Components in an HVAC Control System (Blue lines represent communication.)
2.1 Sensors
Sensors are devices used to produce electrical, digital, or optical signals to measure physical parameters and detect environmental events. Over the past few decades, the sensor and communications market has grown significantly with the development of temperature, humidity, occupancy, light, and CO2 sensors. However, use of these sensors for HVAC control has still not reached its maximum potential in residential buildings for four main reasons: (1) high cost, which can be due to a combination of equipment expense, installation, and maintenance; (2) lack of computational capabilities; (3) inability of HVAC systems to effectively use the data from the sensors; and (4) the need to have sensors at multiple locations. Note that in Table 1 the sensors are listed at the component level. Multiple sensors (along with communication sensors) can be combined together to create products that are of potential use for HVAC systems. Such products are described later in the section. Table 1 provides an overview of available sensors that are either used in existing residential HVAC systems or that are capable of being used in the HVAC systems.
3
On-board sensors and diagnostics are common in HVAC equipment. However, those systems tend to be proprietary, and ultimately help diagnose internal issues with the equipment. Usually, this type of fault helps the equipment be capable of running at all. This paper goes above and beyond the baseline functioning of the system and focuses on ways in which external sensors and controls can help homeowners and technicians save time, money and energy. Therefore, on-board sensors will not be discussed further in this paper.
Table 1. Advantages, Disadvantages, and Applications of Several Types of Sensors
Sensor Types Applications, Advantages & Disadvantages Temperature Applications: Outdoor, rooms, ducts Thermocouples Advantages: Inexpensive. Simple. Self-powered. Large temperature range. Rapid response time. Durable. Disadvantages: Not very accurate or stable. Signal conditioner needed. Thermistors Advantages: Inexpensive. Highly accurate. Stable. Rapid response time. Disadvantages: Fragile. Require voltage excitation. Overheating in certain models. Resistance-temperature Advantages: Long lasting. Accurate. Very stable. Easy to recalibrate. devices (RTDs) Disadvantages: Expensive. Slower response time. Require current excitation. Humidity Applications: Outdoor, dew point, rooms Resistive Relative Advantages: Small size. Low cost. Long-term stability. Field interchangeable. Humidity Sensor Disadvantages: Accuracy is inadequate at low relative humidity RH values. Can shift values if exposed to condensation. Significant temperature dependencies. Exposure to chemical vapors will lead to premature failure. Capacitance Relative Advantages: Stable. Resistant to chemical vapors. Humidity Sensor Disadvantages: Accuracy decreases as temperature deviates from calibration temperature. Field interchangeability requires computer-based recalibration for many models. Distance between sensing element and signal conditioning circuitry is limited to <10 feet. Selection of sensor and transmitter combination is important for proper application. Thermal Conductivity Advantages: Very durable. Resistant to chemical vapors. Absolute Humidity Disadvantages: Accuracy is dependent on ambient temperature. Respond to any gas Sensor that has thermal properties different from those of dry nitrogen. Air Quality Applications: Rooms, ducts Photoionization Detectors Advantages: Accurate. Reliable. Immediate. Can detect multiple types of volatile (PIDs) organic compounds (VOCs). Portable or permanent installation. Can log data. Disadvantages: More expensive than other indoor air quality (IAQ) sensor types. Infrared Carbon Dioxide Advantages: Portable or permanent installation.
Monitors Disadvantages: Acts as a secondary measurement tool for IAQ because CO2 is an indicator of poor IAQ, not a contaminant. Does not measure VOCs from building materials, furniture, and contamination. Metal Oxide Sensors Advantages: Simple. Inexpensive. Can detect multiple types of VOCs. (MOS) Disadvantages: Limited sensitivity. Slower to react. False positives in response to moisture or temperature. Air Pressure Applications: Ducts Variable Resistance Advantages: Simple. Inexpensive. Accurate. Reliable. Durable. Repeatability. Transducers Disadvantages: Amplifier is needed to amplify and condition signal to prevent interference. Changes in temperature may necessitate recalibration.
4
Sensor Types Applications, Advantages & Disadvantages Capacitance Advantages: Can measure small differential pressure changes even with high static pressure. Stable. Repeatability. Disadvantages: High temperature sensitivity. High impedance output. Power Sensors Applications: HVAC equipment Thermistors Advantages: Inexpensive. Highly accurate. Stable. Rapid response time. Disadvantages: Fragile. Require voltage excitation. Overheating in certain models. Occupancy Applications: Indoor occupancy, outdoor motion, garage safety Passive Infrared Sensors Advantages: Simple. Disadvantages: Relies on movement to indicate occupancy. Local in scope. Fairly coarse grained. Ultrasonic Advantages: Can detect motion in any part of a space. Reliable. Typically contain both sensing and circuitry components. Disadvantages: Relies on movement and sound to indicate occupancy. Can be overly sensitive, responding to HVAC airflow.
CO2 Advantages: Inexpensive. Stable. Rapid response time. Disadvantages: Not yet accurate in residential applications. Energy Flow Meters Applications: Metering HVAC, co-generation, water heaters BTU Meter Advantages: Matched sensors make data collection process simpler and maintain data consistency. Output flexibility. Connectivity to common control networks. Disadvantages: Flow measurement accuracy is dependent on model, calibration and installation. Full Bore Magnetic Meter Advantages: Very accurate, even at low flow rates. Long lasting. Reliable. Low maintenance costs. Disadvantages: Expensive. Reverse eddy currents can cause inaccurate readings. Vortex Shedding Meters Advantages: Common. Simple. Inexpensive. Disadvantages: Not accurate at low flows. Ultrasonic Meters Advantages: Noninvasive. Ideal for retrofit applications. Disadvantages: Installation details are critical. Installation error is more common, leading to inaccuracy.
One of the key shortcomings in control technology is that most residential buildings have only one sensor (interior temperature) per HVAC system. This sensor is typically in the thermostat, which is usually centrally located in a home, but does not capture the temperature variations throughout a home. Using multiple temperature sensors could improve both comfort and energy efficiency in homes (Siemann 2010). Similarly, despite the low cost of occupancy (presence/absence) sensors, they are not widely deployed for controlling HVAC systems in residential buildings. Certain thermostats (e.g., NEST, Ecobee) have a built-in passive infrared (PIR) sensor and have been deployed in several million homes. However, their use for efficient, optimized control is still unclear because it is quite challenging to detect occupancy status (occupied/unoccupied) of an entire home based on a single PI sensor at only one location. More examples of sensor applications are described in Section 4.0.
Smart/Intelligent Sensors: “Smart” sensors are sensors with small memory and standardized physical connection to enable communication with the processor and data network. A smart sensor could also be a combination of sensors with signal conditioning, embedded algorithms, and a digital interface. A smart sensor combines a sensing element, analog interface circuit, an analog-to-digital converter, and a bus interface in one unit. It usually provides functions beyond those necessary for generating correct representation of a sensed quantity.
5
Examples of smart sensors that are currently available include digital thermometer with thermocouple probes, multi-meter, infrared sensor/temperature gun, thermo-hygrometer, thermal-imaging temperature gun, and refrigerant pressure gauge. Examples of products that contain smart sensors include iManifold, Comfort Guard, AirAdvice, and FooBot. These products are discussed in more detail in Section 4.0.
2.2 Data Storage and External Signals
Data storage plays a key role in verifying, validating, and improving the commissioning, maintenance, and operation of an HVAC system. Historical data can be processed to learn user behavior and adapt control algorithms accordingly to deliver energy-efficient solutions that improve thermal comfort and IAQ inside homes. There are three options for storing data: (1) on the cloud over the internet, (2) on an on-board device as a part of the existing control system (e.g., smart thermostat), and (3) on a local device that is connected to the existing equipment controllers or thermostats, but not considered to be a necessary part to operate the system (e.g., local gateways and servers).
Several services and products that store data on the cloud are available in the market (e.g., Oracles, Opower). However, these products do not use the stored data for active control purposes; instead they focus on monitoring, diagnostics, and other services that identify potential energy saving measures. Although some devices, like smart thermostats (discussed further in Section 2.6), offer on-board data storage and a little computational power for local control purposes, only some thermostats and some cloudstorage providers offer open interface to access data and control functions. If the connection between these devices is open and easily accessible, the computational power and storage capacity of the entire control system can be significantly increased.
External data signals (e.g., utility pricing, weather, incentive signals) can be considered valuable inputs when deciding the control actions for the HVAC system. If This Then That (IFTTT) is an example of an application that can send external signals to computation platforms like smart thermostats. More use cases for data storage and external signals are discussed further in Section 3.0 of this paper.
Many of the technologies described in this paper incorporate connectedness with the web or the cloud. Consumers may view this connectivity as introducing risk to devices that formerly had none. A report from the California Energy Commission (Ghansah 2012) highlights a few cyber security and privacy issues that are possible by using current programmable communication thermostats and advanced metering infrastructure. The need for cyber security is not exhaustively described in this paper because the issue is ancillary to the actual control functions and applications. However, this need is so fundamental that it is included in the final set of recommendations included in Chapter 5.0.
2.3 Human-in-the-Loop
Human-in-the-loop is the term this paper uses to describe any user’s interaction (e.g., occupants, HVAC technicians, utility operators, etc.) providing input or receiving output from the computation platform. The interface for this interaction can be mechanical (e.g., a physical dial), digital (e.g., a computer screen), or internet-based (e.g., a phone application, tablet, etc.).
One example of a human-in-the-loop is an application that notifies the homeowner when a filter needs to be replaced (e.g., FilterScan). There are also interfaces that a utility control center could use to inform operators of peak events and manipulate many residential HVAC systems at once (e.g., Opower). Those interfaces would likely be part of a much more complex software program for demand response (Somasundaram et al. 2014). These systems could also include third-party vendors that gather data and
6
have a control capability, but offer control services to either a utility or to the consumer (Baechler and Hao 2016).
An example of a human-in-the-loop interface that did not deliver its full potential benefit was the programmable thermostat. According to the Energy Information Administration’s 2009 Residential Energy Consumption Survey (EIA 2009), 43% of heating systems and 48% of cooling systems in residential buildings use a programmable thermostat to control them. Unfortunately, many of the programmable thermostats that have been installed are not properly operated. Of the consumers who own a programmable thermostat, many struggle to properly understand or program its functions. A recent study by Callaway et al. (2014) indicates that programmable thermostats are difficult for users to understand, which has led to ineffective usage or even abandonment. Due to misuse, households with programmable thermostats sometimes have higher HVAC energy consumption on average than those with manual controls. As a result of this finding, the U.S. Environmental Protection Agency (EPA) suspended the ENERGY STAR programmable thermostat program in 2009. The ENERGY STAR program has since initiated a label for certified “smart thermostats” that are proven to save energy compared to a typical thermostat1.
2.4 Communication Between Components
Communication functions allow data transfer between sensors, actuators, controllers, processors, storage, and user interfaces/devices. Communication between these sensors and controls can either be hard-wired, or wireless (local or through the internet). In the direct hard-wired case, a sensor is physically connected using a metal wire, which usually transmits a voltage or current signal, through a computation platform, to a control device and/or actuator. In the local wireless option, sensor data are sent to a control device and/or actuator using a local wireless network. Both the sensor and the control device are required to have a wireless relay installed and configured (e.g., Bluetooth). The internet is not needed in this case. In the internet wireless option, sensor data are transmitted to a control device and/or actuator over the internet. Both sensors and control devices are required to have a wireless sensor installed and be connected to the internet. In this case, the data is usually routed to the device through a server, which can be relatively slow and less reliable compared to any hard-wired connection.
2.5 Computation, Controllers and Actuators
Controls change system operation using both software and hardware. Software receives data from sensors, performs computations, and sends a signal to other external sources. Hardware physically responds to a stimulus control signal. Most control hardware is in the form of an actuator. Historically residential HVAC control systems operated using binary (on/off) actuation. Many newer systems support variable capacity outputs for evaporator and condenser fans as well as for gas burners and compressors. The movement to variable capacity has increased both comfort and efficiency. Common actuators are described in Table 2.
Table 2. Actuation Parameters Used for Controlling the HVAC Systems Actuation Parameters(a) Description A binary signal is sent to turn on or off the cooling mode of operation. This Cooling on/off parameter is usually available for both automatic and manual control in most existing systems.
1 https://www.energystar.gov/products/heating_cooling/smart_thermostats/key_product_criteria
7
Actuation Parameters(a) Description A binary signal is sent to turn on or off the heating mode of operation. This Heating on/off parameter is usually available for both automatic and manual control in most existing systems. The fan is turned on or off, which can be independent of the heating/cooling mode of Fan on/off operation. The damper position is manipulated to change airflow rate in different rooms of a Damper position building. This is not usually prevalent in existing systems. The fan speed can be adjusted in systems with electronically commutated motors to Fan speed change the total supply airflow rate. The compressor speed can be changed in HVAC units with inverter driven Compressor speed compressors. Refrigerant outlet The outlet temperature of refrigerant is changed, which is achieved by changing the temperature outdoor blower speed or compressor speed. (a) Actuation parameters can be changed automatically or manually (by a technician or consumer) during the commissioning, FDD, operation, maintenance, or control stages.
2.5.1 Smart/Intelligent Controllers and Actuators
Smart controllers and actuation devices typically have wider flexibility than traditional controllers and actuators. These devices are capable of integrating themselves easily into HVAC systems (or their smart sensors or intelligent sensing devices) to provide better performance. Examples of smart controllers include variable vent flow devices (e.g., Ecovent), variable speed fans (e.g., Concept 3 motor), and multi- stage refrigeration units (e.g., Carrier’s Infinity® 18VS Heat Pump 25VNA8). Details of these products are discussed further in Section 4.0.
2.6 Integrated Sensors and Controls
Integrated sensors and controls are devices consisting of a combination of sensors and controls that physically receive a signal, do some type of computation, and send a signal to the HVAC equipment. An example of a simple integrated sensor and control is a thermostat (basic, programmable, or smart). A thermostat receives data (mostly room temperature) from sensors and sends an actuation signal to operate the HVAC equipment. Basic thermostats are traditional electrical/mechanical devices that measure temperature and control the on/off state of an HVAC system. Programmable thermostats are digital electronic thermostats with an ability to switch between temperature set points based on the time of day (e.g., unoccupied hours during weekdays) or day of the week (e.g., weekends) as set by the user. Connected thermostats offer enhanced communication, improved interface, and simplified schedule setting (DOE 2016b). Smart thermostats are a type of connected thermostat that contain computation features, and thus are capable of supporting advanced software tools to monitor system performance, detect faults, and ensure correct installation and operation. Smart thermostats (a.k.a learning thermostats,) offer enhanced sensing/communication/computation features, and thus are capable to support advanced software tools to monitor system performance, detect faults, and ensure correct installation and operation. As compared to traditional or programmable thermostats, smart thermostats offer users much more flexibility in the development and implementation of advanced control strategies.
8
3.0 State-of-the-Art and Advanced Applications
In the previous section, components including sensors, data storage, human-in-the-loop, computation platforms, and controls/actuators were each discussed individually. This section describes integrated applications made up of multiple components that provide a service to an end user. These applications are divided into three use-case categories: commissioning, maintenance, and improved operation. In residential buildings, commissioning is likely conducted by the same contractor who installed the HVAC system, or possibly a third-party verifier. This step is sometimes performed as part of the installation. Commissioning can also be performed on existing HVAC systems and equipment, which is referred to as retro-commissioning. An HVAC contractor likely performs maintenance, which is usually initiated by the homeowner, although some homeowners opt for a service contract providing biannual inspections and tune ups. Operation of the system revolves around occupants, and improving this interaction between the occupant and the HVAC system is the subject of Section 3.3. In the future, there would ideally be products that perform tasks across all three categories.
Advanced FDD methods discussed in the sections below can detect faults and prevent catastrophic failures before they happen, thereby saving energy and money while increasing occupant comfort and maximizing the lifespan of the equipment. Advanced controls can use sensor data to operate a system at its optimal performance level based on system specifications, user inputs, and surrounding conditions. Moreover, stored data can also be used to inform and educate occupants and contractors about system operation and design weaknesses, which may result in improved performance by altering the system design or upgrading the system components.
3.1 Commissioning
HVAC commissioning ensures that the installed performance and operation of each component of the HVAC system meet the desired specifications outlined in construction documentation. Ensuring a proper installation requires the following: (1) sensors to measure airflow rate, refrigerant level, refrigerant temperature, zone temperature, etc.; (2) a computation platform to collect the sensors’ data and perform calculations; and (3) a software to process the data and indicate whether the system is installed and configured correctly. These three components together are sometimes referred to as FDD.
Commissioning applications discussed in this section of the report focus on times when an HVAC technician would already be coming to a home, either for initial installation, or for scheduled maintenance (a.k.a. retro-commissioning).
An overview of major commissioning applications, including the current state-of-the-art, the benefit of improving that current technology and the major gaps associated with each application are show in Figure 4.
The detailed analysis of each major application is described in more detail in the subsections below. The tables that discuss the technology gaps of each application include a description of the state-of-the-art technology, potential technology advancements, and a summary of technology gaps. Each table also refers to “inputs,” which include all inputs that help the application complete its task. These can include sensor measurements, manual input/inspection, or stored data from the cloud. The “outputs” refer to all of the outputs that the application provides. These can include a control action, a user notification, or even data sent to the cloud.
9
Figure 4. An Overview of State-of-art, Technical Gaps, and Benefits of Commissioning Applications
3.1.1 Verifying Basic Functionality
In residential buildings, the basic functions of an HVAC system are typically checked by the onsite technician who is installing the system. Checking the basic functionality of the system typically requires a person to turn the system and thermostat on, and visually inspects each of the components. Some of the items a technician would check to ensure basic functionality include: system turns on/off, is each component generally functioning (compressor, fan, burners, sensors, etc.), thermostat signal from equipment. Specific measurements that are taken to determine if there is a fault in the system include: power flowing to equipment, power flowing between equipment components, refrigerant charge, air flow, and connectivity between components. Current state-of-the-art processes and potential technologies are listed in Table 3 below.
10
Table 3. Remaining Opportunities to Reduce Time, Cost and Human Error Associated with Verifying the Basic Functionality of HVAC Systems Description Inputs Output HVAC system Visual inspection NA on/off Components Manual temperature NA Current State-of-the-Art functioning measurement at grills, on-board diagnostics. Thermostat on/off Visual inspection NA Thermostat settings NA NA HVAC system Wireless detection of HVAC Notification sent to on/off system contractor if HVAC equipment is undetectable. Components Current transducers throughout Notification sent to functioning system and temperature sensors. contractor if any power or temperature measurements are out of range. Potential Technology Thermostat on/off Wireless detection of thermostat Notification sent to contractor if thermostat is undetectable. Thermostat settings Wireless timestamp capability. Notification sent to Smart algorithm to determine contractor if HVAC type is HVAC type based on signal, or undetectable. additional current transmitter sensor. Internet-connected current transducers and temperature sensors for remote access of Remaining Opportunity data. Open application programming interface (API) for product connectivity to save time and reduce overall cost.
3.1.2 Duct Leakage and Air Flow
In this application, we first focus on the air flow rate side of the system, which is divided into three parts: (1) detecting any leakage through the duct (and possibly sealing the leakage), (2) ensuring that air flow supplied to a home is close to its designed value, and (3) ensuring that the air flow is well distributed to different parts of the home.
Duct leakage is typically tested using a duct blaster device (e.g., Minneapolis Duct Blaster®). Once leakage is detected in the ductwork, it can be sealed using Aeroseal, which detects small holes and seals them without manual labor for each hole, or more traditional methods if the ducts are accessible. This process is already fairly automated and there are no further controls-related advancements for this particular commissioning application.
For air balancing and total air flow commissioning, an HVAC contractor will typically use a magnehelic gauge with two static pressure probes to test air flow rate near the air handler. A calculation is then conducted to convert the pressure difference into an airflow measurement. A flow hood (e.g., Alnor Flow Hood) that is typically used for measuring the air flow at each register can also be used on the return side to measure the whole-building airflow, but this is not common practice. Variable speed fans can enhance the ability to make automatic changes in the operation if enough air is not supplied to the home.
Balancing air flow rate should primarily be accomplished through good house layout and ductwork design. That said, there is some balancing that can occur through the tuning of grills or manual dampers. There is potential for automatic air balancing and grill tuning using products that are on the market today;
11
however, those products have not been tested by a third party and have not been developed for this specific purpose. Although, more expensive, true multi-zone systems can provide pressure and damper position measurements, which can be used to detect and correct slight balancing issues as well.
Table 4 highlights the largest area for improvement in this application of sensors and controls - the integration of static air pressure sensors into the air handler and duct system. There are few options currently available for inexpensive sensors (both in number and types) available in the market. Permanently installed pressure/airflow rate sensors would provide more accurate and more detailed feedback to installers, who may not be able to fix all of the problems they find with these measurements, but it could help reduce future errors by giving installers the power of data. If permanent sensors were installed, it would also provide an opportunity for a variable speed motor to adjust automatically depending on furniture placement or closed grills, providing better comfort and energy savings.
The open application programming interface (API) connection of any system needs to be managed through some computational platform. One example of a convenient computational platform that could be used for this and other applications is the thermostat.
Table 4. Remaining Opportunities to Reduce Time and Cost Associated with Measuring Duct Leakage and Air Flow Description Inputs Output Automatic duct leakage Duct leakage Digital display and measurement/sealing (e.g., sealed holes Aeroseal) Total air flow measured with Temporary static pressure Digital display Current State-of-the-Art two static pressure probes across probes air handler Measurements taken at each Temperature and pressure Temperature control in register sensors at each register each room Total air flow measured with Permanent pressure Automatic adjustment two static pressure probes across probes/gauge of variable fan speed air handler to match total air flow Potential Technology Auto-balancing of air-side Permanent pressure probes Auto-balanced air-side system in air handler and at of HVAC system registers
Permanent pressure/airflow rate and temperature sensors for better feedback to the Remaining Opportunity installer and automatic adjustment of variable speed motor. Open API for product connectivity to save time and reduce overall cost
3.1.3 Refrigerant Charge
In this application, sensors detect problems with the refrigerant side of an HVAC system—such as overcharging, undercharging, and leakage—which can be immediately resolved onsite.
The most common way to measure refrigerant charge during the commissioning process is to use a temporary data acquisition system (e.g., iManifold, Testo, Sporlan). Some products integrate an actuation into the device, such as the “Mastercool 99661-A Digital Manifold Gauge Set,” which can automatically charge refrigerants to an optimized level. These products (1) detect the pressure of the refrigerant, (2) automatically stop the charging when the desired set point is reached, and (3) continuously transmit the
12
information to a digital display (wired or wireless) or to the technician’s smart phone or tablet during the entire process.
The potential technology described in Table 5 includes a set of permanent manufacturer-installed sensors in the outdoor unit of a system that uses refrigerant. The benefit of permanent sensors and chargers is that it could save time and money for future applications. Internet connectivity of sensors like this would provide an opportunity for remote quality assurance.
Table 5. Remaining Opportunities to Reduce Time, Cost and Human Error Associated with Charging Refrigerant Description Inputs Output Temporary data collection Refrigerant temperature and Digital display, auto- and automatic shutoff of pressure shutoff when desired Current State-of-the-Art charging system (e.g., pressure is reached Mastercool 99661) Permanent sensor Refrigerant temperature and Digital display, auto- installation with automatic pressure shutoff when desired Potential Technology flow shutoff when fully pressure is reached charged Permanent refrigerant gauge and charger to capitalize on opportunity and save money Remaining Opportunity for future use. Internet connectivity of sensors like this would provide an opportunity for remote quality assurance.
3.2 Maintenance
Maintenance is necessary to ensure the ongoing performance of HVAC systems, equipment, and related components. Preventative maintenance (a.k.a. retro-commissioning) focuses on tune ups or other actions that are based on a timeline, rather than a physical trigger or fault. Reactive maintenance is based on a physical measurement or fault that signals that something is not performing as it should. Herein, reactive maintenance is referred to simply as maintenance in this report.
As discussed in Section 3.1, FDD can be used to identify, isolate, and correct or avoid faults that develop over time in existing HVAC systems. Although there is overlap between preventative maintenance and reactive maintenance, this section focuses on reactive maintenance, where permanently installed (and internet-connected) sensors could help reduce the time and cost associated with unexpected HVAC faults.
An overview of major maintenance applications, including the current state-of-the-art, the benefit of improving that current technology, and the major gaps associated with each application is shown in Figure 5. The detailed analysis of each major application is described in more detail in the subsections below.
13
Figure 5. An Overview of State-of-art, Technical Gaps, and Benefits of Maintenance Applications
3.2.1 Verifying Basic Functionality
The basic functionality of the system can be monitored and corrected using the same approach as is discussed in Section 3.1. In this case, data storage may be a helpful addition to the gaps previously mentioned. Data storage either locally or on the cloud could help determine if measurements are changing over time. This type of data could help detect a faulty sensor/sensor drift or even a dying motor.
In case of faulty sensors, an appropriate action (alert, control action, etc.) can be taken to compensate for inaccurate sensors so that the system performance is minimally affected. Examples include removing a faulty sensor from control system architecture, and changing the temperature set point if the humidity or temperature sensor is offset by a known value. Other applications include checking thermostat
14
settings/parameters, data reporting, and verifying the sequence of operations and control functionalities. The requirements and potential advancements are similar to the ones discussed in Section 3.1.
3.2.2 Refrigerant Level
This application is similar to the “refrigerant charge” application described in Section 3.1. The refrigerant level is a key factor in HVAC performance and energy efficiency. This includes overcharging, undercharging, and leakage over time.
There are currently no permanently installed, internet-connected refrigerant sensors or chargers, with an open API available on the market. The benefit of having permanently installed refrigerant temperature and pressure sensors is that faults could be detected quickly. If those sensors are internet-connected, they could also immediately notify a contractor or homeowner that a fault is about to hinder a system from functioning efficiently, or functioning at all. ComfortGuard is a proprietary package of sensors that must be installed by a certified HVAC technician. This is the only product on the market that might contain internet-connected refrigerant sensors, but this is not information that is shared publically. If products like this were available with an open API, they could connect with endless third-party apps, thermostats and even large data collection software for research purposes.
A refrigerant canister/charger could also provide automatic re-charging of refrigerant systems to minimize or eliminate the effect of refrigerant leaks on equipment performance.
Table 6. Remaining Opportunities to Reduce Time and Cost Associated with Checking and Changing the Refrigerant Level Description Inputs Output Temporarily installed Refrigerant temperature and Digital reading and cloud sensors with data storage pressure storage of measurements in cloud (e.g., iManifold) Current State-of-the-Art Permanently installed Refrigerant temperature and Notification to user sensors (e.g., pressure Comfortguard) Permanent and internet- Refrigerant temperature and Wireless notification to connected refrigerant pressure users. Automatically Potential Technology sensors and chargers with adjust variable speed fan open API if applicable. Permanent, internet-connected refrigerant sensors for remote access. Open API to save Remaining Opportunity time and reduce cost.
3.2.3 Air Filter Replacement
Air filter replacement is an important maintenance task that is usually recommended once every two or three months. Homeowners are typically responsible for filter replacement, and it is common for people to forget about it. The DOE (DOE 2016a) describes how replacing a dirty, clogged filter with a clean one can lower your air conditioner's energy consumption by 5% to 15%. In addition, the EPA (EPA 2016) mentions that incorrect air flow could waste up to 3% of HVAC energy use. Most smart thermostats will provide a time based notification to replace filters. A product like FilterScan could help detect dirty filters, but also might be used for general airflow maintenance if the sensor data was available through an open API. This product detects change in differential pressure to detect whether the filter needs replacement. Then, a notification (text, email, etc.) is sent to the user to replace the filter. An
15
improvement on this technology would be if it were integrated into the air handler upon installation. If an open API were available, these types of sensors could potentially communicate with a variable speed motor, which could always be adjusting to the airflow available and run the fan as efficiently as possible. This application is summarized in Table 7.
Table 7. Remaining Opportunities to Reduce Time and Cost Associated with Air Flow Issues in HVAC Systems Description Inputs Output Detection in the change in Pressure differential Notification to pressure difference across homeowner Current State-of-the-Art air handler (e.g., FilterScan) Detecting dirty filters Pressure differential Notification to (also their location) and homeowner. Alter Potential Technology changing the air flow rate variable speed fan for to compensate for dirty optimum performance. filters Permanent and internet-connected air pressure sensors that are integrated into all air Remaining Opportunity handlers and automatic adjustment of variable speed motor through an open API
3.2.4 Less Common Faults
This section includes all other faults that were not included in previous applications such as dirty coils, condenser drain, worn out parts, lubrication, faulty sensor reporting, communication delays, etc. Some of the state-of-the art products that are available on the market today are discussed here. There are many sensors that HVAC technicians can use once they are on site. The drawbacks of temporary sensors are the time required to install them and the potential for human error. The products listed in this section are permanently installed to help ensure accurate diagnostics and quick response times.
Some connected thermostat products (e.g., Nexia Diagnostics, Ecofactor) are able to determine some faults (incorrect/swapped wire installation, tripped breakers, etc.) of the HVAC system, simply through the thermostat signals. A full discussion of the use cases for connected thermostats is available in a recently published report (DOE 2016b).
ComfortGuard uses 10 sensors and monitors the system performance in real time. It also predicts system performance and provides diagnostic alerts (to homeowner or contractor) regarding dirty filters, heating/cooling efficiency, and component-level performance such as flame sensors, hot surface igniters, capacitors, pressure switches, thermal limits, coil effectiveness, expansion devices, refrigerant charging levels, reversing valves, defrost boards, fan motors, compressors, and contactors.
All of these applications of sensors and controls are helpful in FDD for residential HVAC systems. However, they are all relatively expensive and require homeowners to purchase the whole set of sensors from one manufacturer. It would be ideal if these products were available separately, integrated into the air handler and/or outdoor unit during manufacturing, and connected to the internet for data collection and communication.
16
3.3 Improved Operation
An HVAC system is operational if it is correctly responding to a thermostat and delivering air within manufacturer specifications. Basic HVAC performance can be improved upon in three ways: (1) comfort, (2) energy savings, and (3) grid power management.
The sub-topics below are divided into these three improvement categories, although, many of the sub- topics could apply to multiple improvement areas. For example, applications in the energy savings group are primarily focused on reducing energy consumption although they may also lead to improved comfort.
An overview of major improved operation applications, including the current state-of-the-art, the benefit of improving that current technology, and the major gaps associated with each application is shown in Figure 6. The detailed analysis of each major application is described in more detail in the subsections below.
Figure 6. Summary of Improved Operation Applications
17
3.3.1 Operation Applications Focused on Comfort
3.3.1.1 Temperature Set Point
Thermostats provide temperature set point control. Basic thermostats measure temperature and use a manually input set point (analog or digital) to send a “turn on” or “turn off” command to the HVAC system. Programmable thermostats have the ability to change temperature set point based on a pre- defined user-input schedule. Connected thermostats offer enhanced communication, improved interface, and simplified schedule setting (DOE 2016b). Smart thermostats are a type of connected thermostat that contain computation features, and thus are capable of supporting advanced software tools to monitor system performance, detect faults, and ensure correct installation and operation.
Smart thermostats offer the most advanced way of controlling the indoor temperature. With smart thermostats, the temperature set point can be optimally controlled based on occupant behavior, outside temperature, historical set points, energy cost, or utility signals. The optimal set point can be calculated over the cloud or using the computational power available on the device. Depending on control algorithms, measurements from several sensors and data from users may be needed. The combination of two-way communication, data collection, and modeling human behavior has made these devices a platform for vendors to offer demand-response services to utilities (Baechler and Hao 2016).
Smart thermostats (e.g., NEST, Ecobee, etc.) are currently the state-of-the-art application of temperature set point control. EnergyStar has added a smart thermostat category1, which can help identify a certain level of energy savings that can be expected from some thermostats. However, other than this Energy Star distinction, the amount of comfort gained and energy saved across the spectrum of smart thermostats available is different and currently cannot be easily compared.
3.3.1.2 Ventilation, Filtration and Indoor Air Quality
Ventilation provides fresh outdoor air to the spaces inside buildings. The primary function of ventilation control is to adjust the airflow (determined by standards like ASHRAE 62.2) entering or exiting a building. Inputs into this control task can be as simple as the time of day, or can be based on factors such as CO2 measurements, outdoor temperature/humidity, and occupancy. Filtration can occur with or without ventilation and typically uses physical barriers to minimize particulates in the air that is circulated throughout a home. Both of these mechanisms can be associated with better IAQ (Singer et al., 2017). Table 8 focuses on how ventilation and filtration can be used to optimize IAQ and energy use.
Table 8. Missed Opportunities to Improve IAQ while Minimizing Energy Use Description Inputs Output ERV/HRV(a) Time of day, manual On/off state of motor Whole house air NA, Ultraviolet (UV) light UV light purification always on Current State-of-the-Art IAQ sensors that have CO2, VOCs, PM, CO, Air handler on/off state, open API and can control temperature, humidity and user notification a thermostat (e.g., Foobot) Smart ventilation: Formaldehyde, PM, CO2, Automatically changes Potential Technology optimizing input from CO, VOCs, outdoor ventilation rate and turn IAQ sensors to control conditions, number of on/off filtration system;
1 https://www.energystar.gov/productfinder/product/certified-connected-thermostats/results
18
Description Inputs Output ventilation and filtration occupants user notifications system (perhaps at variable rates) Smart ventilation integrated with filtration system. Integration with a variable speed Remaining Opportunity air handler through an open API would provide opportunities for energy savings by matching air flow rate to pollutant level. (a) ERV and HRV are energy recovery ventilation and heat recovery ventilation, respectively.
Smart ventilation is gaining momentum in research and in industry. The goal of some smart ventilation systems is to save up to 40% of ventilation related energy use (Walker, 2014). However, a major gap is the ability to control ventilation rates in all types of ventilation systems, using some of the same sensors that are used for other applications (but that may use proprietary communication protocols).
3.3.1.3 Humidity
As homes continue to have a lower and lower HVAC load, humidity becomes a more and more important factor to address directly (Brown et al., 2013). State-of-the-art humidity control devices can be integrated into the main air handler unit in a home. These humidifiers and dehumidifiers provide the ability to meet a certain humidity set point that is set by the user at the thermostat (or in a thermostat app).
Humidification and dehumidification are energy intensive processes due to the phase change of the liquid to vapor or vice versa. Therefore, any opportunity to save energy during this process should be capitalized on. Table 9 describes the missed opportunities to save energy when humidifying or dehumidifying a home.
Table 9. Missed Opportunities to Save Energy Associated with (De)Humidification Description Inputs Output Whole house Temperature and humidity Opens/closes water valve, (de)humidification or turns on the compressor Current State-of-the-Art and dehumidification blower set point to meet humidity set point Variable rate Indoor and outdoor humidity Changes valve position, (de)humidification and occupancy sensors that compressor and blower Potential Technology with open API detect number of people or speeds to meet humidity set activity level point Variable rate (de)humidification could react to indoor and outdoor conditions and Remaining Opportunity deliver ideal humidity levels with the least amount of energy use. Open API would enable connection to existing or third-party occupancy sensors or other measurements.
3.3.2 Operation Applications Focused on Energy Savings
3.3.2.1 Natural Heating and Cooling
This application uses natural heating and cooling if outdoor weather conditions are favorable. Night ventilation and operation of windows are two ways to accomplish natural heating and cooling in residential buildings. To implement this application automatically, however, there are two major requirements:
19
actual outdoor weather measurements and forecasts, which can be obtained using weather website APIs or local weather stations/sensors that send data to the cloud the ability of an HVAC system to allow natural heating or cooling, (e.g., supply/exhaust fan control, outside damper control, automatic window control), which is not common in currently available systems.
Table 10 shows the current state-of-the-art application and how it might be improved upon for the best possible control of natural heating and cooling.
Table 10. Missed Opportunities to Save Energy with Natural Heating and Cooling Description Inputs Output Automatic ventilation Outdoor temperature Opens/closes damper or and based on favorable outdoor and humidity turns fan on Current State-of-the-Art air conditions (e.g., NightBreeze) Fully integrated and Indoor and outdoor Changes damper position and automated natural temperature and fan (exhaust, supply) speed Potential Technology ventilation with high humidity Opens/closes windows control options Automatic natural ventilation during favorable outdoor and indoor conditions using an Remaining Opportunity open API to connect to the main HVAC system.
3.3.2.2 Multi-Zone Temperature Distribution
In this application, different HVAC zones can be created and controlled independently by using additional actuation sensors and hardware at multiple locations. More than 70% of residential buildings use one thermostat (RECS 2009)1 and are not capable of controlling the environment in individual rooms independently. Although most OEMs offer a multiple zone system, they are very expensive and typically only practical in new construction. A less expensive implementation of a zoned system at a large scale requires the following: low-cost temperature/humidity sensors that can be easily integrated into existing HVAC system. Some sensors that are available today can be connected to a thermostat through open API interfaces like IFTTT, but the full integration with the thermostat to create a zoned home based on sensor feedback has yet to be tested. These sensors currently cost $25-60 per sensor. inexpensive diffusers or dampers to control the amount of air flow rate entering different rooms. Although several commercial products (Ecovent, Keen Home Vent, etc.) are available for retrofitting the exiting diffusers, the products are relatively new and only being deployed primarily to enhance comfort. development and implementation of control algorithms. Currently, smart diffusers control the damper positions based on temperature set points, which can be set through occupant’s smart phone apps. These apps have limited control commands based on mostly temperature settings in a home. More advanced controls with multiple sensor inputs (e.g., occupancy and temperature) have not been tested.
To capitalize on the energy savings benefit related to multiple zones in a house, the air flow should be dynamically adjusted according to the registers available. Reducing the fan speed lowers both the fan
1 Although more detailed questions about HVAC equipment have been added to the RECS database since 2009, the question regarding the number of thermostats in a home was not reported on after 2009.
20
energy and heating/cooling energy consumption. Although there are commercially available products with more than one fan speed (e.g., INFINITY 18VS heat pump 25VNA8, Goodman 3 Ton 16 SEER AC Model DSXC160361 with Goodman Multi-Position Air Handler with Variable Speed Fan Model AVPTC48D14), they are not prevalent in existing buildings. Therefore, in this application, air handler or air handler motor replacement would be recommended (to enable variable speed) along with the multi- zone control.
Improved comfort is also a benefit of this application. If a home experiences high temperature differences between certain parts of the home, or has multiple stories, more uniform temperature distribution may be possible with variable air flow rate supplied to each room.
Table 11 describes the current state-of-the-art multi-zone systems and what kind of product might be an improvement on the state-of-the-art system.
Table 11. Remaining Opportunities to Save Time and Cost When Installing Multi-Zone Systems Description Inputs Output Wireless control of room Temperature and Varies damper position for each grill dampers based on varied pressure set points (e.g., KEEN), retrofit solution Current State-of-the- Wired control of zone Temperature Varies damper position for each zone Art dampers based on varied set points (e.g., Honeywell Wireless TrueZone), hard- wired solution Automated control of Temperature, air Adjusts damper and fan speed dampers and air flow rate flow rate and Potential Technology using other sensors (e.g., occupancy occupancy) Remaining Integration of multi-zone systems with other sensors and variable speed equipment with Opportunity open API.
3.3.2.3 Complete Smart HVAC Optimization for Energy Reduction, Health and Comfort
This application uses predictive and optimization control approaches to reduce energy consumption, while maintaining IAQ and comfort in a specified range. Implementing this application requires much higher computational power than other applications. Therefore, it is likely that all of the computation would be performed on the cloud and the control actions can be sent through the internet. Depending on the type of control approach, the application may need weather forecast, occupancy behavior, and other models for predicting the system behavior and taking an appropriate action accordingly. This type of system would include control over the entire environment of the home, including temperature, humidity, multi-zoning, automatic shading, and mechanical and natural ventilation. Table 12 describes the current components that exist for this type of system, as well as the remaining opportunities in this fully integrated application.
Table 12. Remaining Opportunities for Complete HVAC Optimization for Energy Reduction, Health, and Comfort Description Inputs Output
Current State-of-the-Art Complete system from one CO2, VOCs, PM, CO, Controls temperature, humidity,
21
Description Inputs Output company (e.g., Honeywell occupancy, ventilation in a sort-of Whole Home, Lennox temperature and independent fashion with Complete iHarmony etc.) humidity limited data exchange Disjointed retrofit products CO2, VOCs, PM, CO, Separate control of HVAC (e.g., Foobot, Ecovent, occupancy, system components including smart phone, occupancy temperature and set back, multi-zoning, sensors, smart thermostat, humidity automatic shading, mechanical Keen) that provide input to and natural ventilation. different computation platforms Integrated sensor products CO2, VOCs, PM, CO, Optimized control HVAC that easily connect to a air flow, occupancy system including set back, Potential Technology central hub which can and indoor/outdoor multi-zoning, mechanical and optimize a complete HVAC temperature and natural ventilation. system humidity Energy savings and comfort associated with a complete smart system. Ease of Remaining Opportunity connectivity and optimization between components and an improved user interface would enable more people to use such a product.
3.3.3 Operation Applications Focused on Grid Services Integration
HVAC systems can be used to provide grid services, which may not necessarily result in reduced energy consumption. There are several applications in this area: (1) integration and coordination of multiple HVAC systems and other components at the community level, (2) enabling automated demand response through individual HVAC system control, (3) using residential buildings as a micro-grid and coordinating among the generation resources and loads inside residential buildings, (4) precooling to shift or shave the peak load, (5) duty cycling the HVAC unit, and (6) deciding the time interval between on and off times can be used for reducing peak load. All of these services can be provided through smart thermostats and are available with the technology on the market today. Discussion of the use cases for smart thermostats is available in a recently published report (DOE 2016b).
22
4.0 Summary of State-of-the-Art Product Features
This section provides examples of the connected residential HVAC products currently available. State-of- the-art component features are listed in tables, which help provide a picture of what advanced applications might be achievable using just the products on the market today. Automating each application listed in Section 3 would require each of the physical component technologies (sensors, data storage, human-in- the-loop, and controls) to be connected. This exercise has revealed that besides thermostats, most of the products on the market today focus on one end of the spectrum (sensors) or the other (controls), but not both – therefore connectivity between products is a key component that cannot be ignored.
The tables do not contain all of the products on the market, only examples of products that have many desirable features. The products are organized by application type, then by major product category (e.g., refrigerant-related products). These tables do not include manufacturers who just provide software because the software itself does not sense or control anything. The term “open API” that is used in the tables below refers to the ability for the product mentioned to connect to at least one other product via an API. To get an “x” in this category, the product does not need to have an API that is open to the public.
Table 17 shows the products in the thermostat category. The features captured here include all features that are possible with that thermostat and potentially some thermostat add-ons (like humidity relay switches), but does not include any capabilities that would be purchased from a separate company and added to the base product.
Table 13. Examples of Commissioning and Maintenance Refrigerant-Related Product Capabilities
No
–
information available) is public
Product Name Detection and Diagnostics
ComfortGuard (Connected Fault
iManifold Testo SmartProbes ManTooth Dual Pressure Wireless Digital P/T Gauge SmartService ToolKit Sporlan Features
Do-It-Yourself Install Sensors Air Temperature x
Humidity x
Air Pressure x
Refrigerant Temperature x x x x Refrigerant Pressure x x x x
Data Storage and External Signals
Able to receive external signal from utility
23
No
–
information available) is public
Product Name Detection and Diagnostics
ComfortGuard (Connected Fault
iManifold Testo SmartProbes ManTooth Dual Pressure Wireless Digital P/T Gauge SmartService ToolKit Sporlan On-board data storage x
Cloud data storage x
Human-in-the-loop User Programming
Automated Programming
Scheduled alerts
Automated alerts x x
Communication Between Components Wi-Fi
Bluetooth x x x
Open API
ZigBee x
Controls Heating/cooling on/off
Humidity level
Ventilation
Fan on/off
Damper position
Fan speed
Zone control
24
Table 14. Examples of Air-Side Commissioning and Maintenance Product Capabilities
Product Name
FilterScan Honeywell Ventilation Systems
Features
Do-It-Yourself Install x Sensors Air Temperature x
Humidity x
Occupancy
CO2
Air Pressure x x
Data Storage and External Signals On-board data storage
Cloud data storage
Local server
Human-in-the-loop User Programming
Automated Programming
Scheduled alerts
Automated alerts x
Communication Between Components Wi-Fi x
Bluetooth
API to Connect with Thermostats
ZigBee
Controls Heating/cooling on/off
Humidity level
25
Product Name
FilterScan Honeywell Ventilation Systems Ventilation
Fan on/off
Damper position x
Fan speed x
Zone control
Table 15. Examples of Advanced Thermostat (Operation) Product Capabilities
Examples of Original Equipment Manufacturer
Examples of Aftermarket Thermostats Thermostats
Gen)
nd
4
Lyric
Fi
(Nexia)
-
Sensi
Wi
Thermostat E
ComfortLinkII
Ecobee
WiserAir
ControlCenter
Rheem EcoNet
Lux Kono PRO
XL1050
Honeywell
Lennox iComfort
NEST Trane
Product Name Thermostat (2
Carrier Infinity Touch
Caleo (baseboardheat) Features
ENERGY STAR x x
Do-It-Yourself Install x x x x x x x
Sensors Temperature x x x x x x x x x x x Humidity x x x x x x x x x x Occupancy x x
Light x x
Data Storage and External Signals Able to receive external signal x x x x x from utility On-board data storage x
Cloud data storage x x
26
Examples of Original Equipment Manufacturer
Examples of Aftermarket Thermostats Thermostats
Gen)
nd
4
Lyric
Fi
(Nexia)
-
Sensi
Wi
Thermostat E
ComfortLinkII
Ecobee
WiserAir
ControlCenter
Rheem EcoNet
Lux Kono PRO
XL1050
Honeywell
Lennox iComfort
NEST Trane
Product Name Thermostat (2
Carrier Infinity Touch
Caleo (baseboardheat) Geofencing x x x x x x
Human-in-the-loop User programming x x x x x x x x x x x Scheduled alerts x x
Automated alerts x x x x x x x
Energy reporting x x
Integrated voice control x
Communication Between
Components Wi-Fi x x x x x x x x x x Bluetooth x
Open API x x x x x x
Zigbee x
Controls and Actuators Heating/cooling on/off x x x x x x x x x x Humidity level x x x x x x Ventilation x x
Fan on/off x x x x x x x x x Fan timing x x x
Multi-zone (control more than one x x thermostat) Dual fuel capable x x
2 or more stage equipment x x x x x x
Variable capacity equipment x
Heat pump with auxiliary heat x x x x
27
Table 16. Examples of Indoor Air Quality (Operation) Product Capabilities
ot
Product Name
Foob NetAtMo Awair Breezometer FilterScan AirAdvice Honeywell Ventilation Systems NightBreeze Aprilaire Dehumidifier
Features
Do-It-Yourself Install x x x x x x Sensors Air Temperature x x x x x x
Humidity x x x x x x
Occupancy
CO2 x x x x x
Air Pressure x x
Data Storage and External Signals Able to receive external signal from utility
On-board data storage x
Cloud data storage x x x
Local server x
Human-in-the-loop User Programming x x
Automated Programming
Scheduled alerts
Automated alerts x x x
Communication Between Components Wi-Fi x x x x
Bluetooth
API to Connect with Thermostats x x x
ZigBee
Controls Heating/cooling on/off
28
ot
Product Name
Foob NetAtMo Awair Breezometer FilterScan AirAdvice Honeywell Ventilation Systems NightBreeze Aprilaire Dehumidifier Humidity level x
Ventilation x
Fan on/off x
Damper position x x
Fan speed x x
Zone control
Table 17. Examples of Other Advanced Operation Products and Product Capabilities
(Integrated
o Weather Station
Product Name
EcoVent KEEN Home Vent Honeywell Complete System) HVAC NetAtM Lennox iHarmony Symaro Sensors
Features
Do-It-Yourself Install x x x x Sensors Air Temperature x x x x x
Humidity x x x
Occupancy
Air Pressure x x x x
Indoor Air Quality x
Sound x
Data Storage and External Signals Able to receive external signal from utility
29
(Integrated
o Weather Station
Product Name
EcoVent KEEN Home Vent Honeywell Complete System) HVAC NetAtM Lennox iHarmony Symaro Sensors On-board data storage
Cloud data storage
Local server x x
Human-in-the-loop User Programming x x
Automated Programming x x
Scheduled alerts
Automated alerts
Communication Between
Components Wi-FI x
Bluetooth
ZigBee x
Open API x x x x
Controls Heating/cooling on/off x
Humidity level x
Ventilation x
Fan on/off x x
Damper position x x x x
Fan speed x x
Zone control x
30
5.0 Remaining Opportunities and Barriers
This section summarizes the remaining opportunities that could help decrease time, cost, energy use, and human error associated with various advanced HVAC applications. The technical aspects of this section are organized based on the components described in Section 2: sensors, data storage and external signals, human-in-the-loop, communication between components, and finally, computation, control, and actuation. Market barriers round out this section to capture the reoccurring non-technical themes that appear throughout the paper.
The studies mentioned in this section of the paper use different baseline controllers, buildings, and systems. Therefore, these studies cannot be compared in a unified manner to conclude anything. Confounding variables from each study ensure that the findings from individual studies cannot be added up to find a meaningful comprehensive result. The studies are merely presented to show that advanced sensing and control can result in significant energy savings and improved comfort.
5.1 Sensors
Occupancy sensors have the potential to save a lot of energy in homes (Piper et al., 2017). From setting back a thermostat, to lowering the ventilation rate, there are a number of applications that could benefit from accurate and inexpensive occupancy sensors. There are a few mechanisms by which occupancy is currently measured in homes. Motion sensors, GPS/geofencing locators, and PIR sensors can be inexpensive ways to measure occupancy; however, these methods have some limitations. Advanced occupancy sensors have the potential to measure the number of people and their activity level in homes. This type of advanced sensor research is currently being done by the Department of Energy’s ARPA-E Program1 and therefore will not be discussed further.
Other sensors have recently become much more accessible. Motion, temperature, and humidity sensors are now internet-connected, are relatively inexpensive ($30-$60 each), and can connect to other devices through open API apps such as IFTTT (e.g., D-Link Motion, ThermoPeanut, MESH, etc.). No known research has been done with the latest products to determine the level of accuracy or the potential calibration offset over time. If these sensors have reasonable accuracy over time, they could be used as conduits for other measurements. For example, with more research a temperature sensor could potentially be attached to the outside of a refrigerant line to help monitor the refrigerant temperature over time to determine if there is a sudden change in the temperature of the refrigerant, which often is associated with certain faults. More research related to the use of low-cost sensors in a home could be helpful to determine possible applications for commissioning, maintenance, and operation.
5.2 Data Storage and External Signals
Storage and accessibility of equipment meta data could help reduce errors, time on site, and ultimately make maintenance of HVAC systems less expensive. For example, every time a technician visits a site, system information (e.g., refrigerant type, stages, equipment type) could be entered for diagnosis and maintenance purposes. Automatic availability of such information to contractors who connect to the system (through Wi-Fi for example) could reduce errors, decrease time spent on the site, and expedite the overall process. Availability of such data (including equipment characteristics) can be quite helpful in detecting an outlier/fault or optimizing the system performance during operations in automated fashion.
1 https://arpa-e.energy.gov/sites/default/files/documents/files/SENSOR_ProgramOverview_FINAL.pdf
31
Moreover, advanced controls could also achieve a significant amount of energy and cost saving when responding to a grid dispatch signal for demand response purpose. For example, by responding to a real- time price signal from the grid, the average electricity energy saving due to peak load shaving is about 5% (OPOWR 2014; Yoon et al. 2014; Consumers Energy 2014).
5.3 Human-in-the-Loop
Consumer interaction plays a significant role in the efficient operation of an HVAC system. Several field deployments have shown that consumer feedback devices can lead to 12-15% energy savings (GridTalk (2013) and Parker et al (2006).
The inability to easily incorporate the end user (actively or passively) in the control loop can have the reverse effect. A recent study by Callaway et al. (2014) indicates that programmable thermostats of the time were particularly difficult for users to understand, which led to ineffective usage or even abandonment. Programmable thermostats capture over 58% of the market, yet 45% of households with thermostats still set one temperature and leave it most of the time (EIA 2015).
Recent smart thermostat products seem to address this issue by creating an interactive and flexible user interface. These products seem to be gaining popularity in the market, although as of 2015, the market share was still only 4% (EIA 2015). At this point, although the number of smart thermostat users is relatively low, one study shows that those users are in fact energy conscious and are using the thermostats to their full potential. A study by Apex Analytics (2017), using NEST thermostats, showed that about 80% of the people who use them were interested in saving energy/money1. As these thermostats become more mainstream, it will be important to continue to conduct research about if users are allowing the technology to reach its full potential.
Another area for user interface improvement is related to the ease of use of combined components. Ideally, one interface would exist that would control and optimize all HVAC-related components. This software/hardware could use a home’s Wi-Fi connection to automatically detect and connect all of the components to a simple user interface. Potential areas of interest are summarized below: development of an interface and learning algorithms that maximize sensor input and simplify direct user interaction to decrease the occupant burden during setup and use. an inexpensive and user-friendly hardware and software platform that is compatible with all components. an interface that consumers can use to set acceptable utility control parameters for demand response programs.
5.4 Communication Between Components
Easy implementation of advanced control applications on a large scale would require a high level of compatibility between equipment from different manufacturers, especially when multiple systems, such as heating, cooling, humidification, dehumidification, and ventilation systems are involved. The use of open standards for residential controls would allow for more ready integration of controls. Currently, most OEM solutions utilize proprietary communications with the exception of the wiring standard to connect thermostats.
1 Users were asked to opt-in to an energy saving program and about 80% of users accepted. Attrition was minimal and was less than 5% at the conclusion of the study.
32
Smart or connected thermostats play a crucial role in connecting and interacting with HVAC related equipment and systems. These thermostats provide hardware to gather user inputs, compute control actions, and send actuation commands to HVAC equipment. Other state-of-the-art computation platforms are also entering the market. With multiple manufacturers responsible for the HVAC equipment, user interfaces, sensors, and actuators, there is a growing market for open API communication protocols that enable interoperability among these technologies.
Most manufacturers do not provide the capability to extract data from the system. However, they will allow a connection to certain other products in order to improve the flexibility and functionality of their product. For example, some manufacturers are working with each other directly (e.g., Ecobee and Amazon Alexa). Even more manufacturers are signing up for a more open integration platform, like IFTTT, which can use hundreds of devices to sense and control various products in a home. It would be beneficial to have all sensors, thermostats, actuators, and HVAC equipment with the following capabilities: ability to incorporate additional sensors into the system seamlessly, including “plug and play” and auto-configuration functionality. availability of component-level information (including sensor data), which may require establishing two-way communication between devices. cyber security to prevent unauthorized access of connected equipment.
In general, there are a few product categories that do not currently have any compatibility or communication capabilities. For example, most refrigerant-side monitoring equipment does not connect to Wi-Fi, nor does it connect to other devices in the home. Although this equipment is currently used as a one-time commissioning or retro-commissioning product, with enhanced communication capabilities, it could be used for maintenance as well. Other product categories with little to no compatibility with other vendors include (de)humidification equipment and ventilation equipment.
5.4.1 Opportunities for Refrigerant Sensor Communication
A glaring missed opportunity related to communication between components appears in Table 15 (Section 4.0). This table shows how most of the products on the market today that could help detect and communicate faults from the refrigerant side of heat pump and air-conditioning systems are not permanently placed, nor are they connected to the internet.
Goetzler et al. (2012), show that central AC and heat pump FDD algorithms could save 0.40 quads of energy. In particular, improper refrigerant charge is the most important fault to be addressed. Proctor (2004) reported that about 62% of residential air conditioners had incorrect refrigerant charge, and 35- 45% of residential HVAC systems suffer from insufficient refrigerant charges; this decreases the system efficiency by 12-20%. Detecting low charge faults in the residential AC units and correcting them would reduce the cooling energy consumption by 5-10%, which is equivalent to 0.1-0.2 quads of energy savings (Roth et al. 2006). FDD provides several other significant benefits that also amortize over the life of an HVAC system. These benefits, beyond reduced electricity use, include more efficient preventative maintenance services, increased system lifetime, reduced number of repair parts, and reduced labor (Roth et al. 2006; Wiggins et al. 2012).
33
Clearly, permanently placed, internet-connected refrigerant sensors with an open API could be used to help save a substantial amount of energy.
5.4.2 Automated System Zoning
Another example of interconnectivity that could provide a substantial amount of energy savings is automated system zoning. This would require sensors (e.g., temperature, occupancy, and potentially air pressure) to be connected to controllers (e.g., air dampers) through a central hub that could perform computations and interaction with the user. Brown et al. (2007) used an occupancy schedule to dictate the set points for individual zones during the heating season. The report shows 26% energy reduction with register controls. A multi-zone setback strategy implemented by Oppenheim et al. (1992) results in 12% energy savings as compared to a single-zone strategy with an 8-hour 12°F setback. Automating this type of control could help ensure that occupant comfort was not sacrificed for energy savings.
5.4.3 Opportunities for Ductless Mini-split Communication with Central Systems
For the last 5 years or so, many utilities have provided incentives for residential ductless mini-splits, with most homeowners adding one indoor unit in a central living area. Controls for these systems have been disjointed and not suitable for pairing ductless mini-splits with pre-existing systems such as forced-air furnaces, zonal electric resistance, or boilers with radiant heaters. One study has shown that the interaction between ductless mini-splits and central systems is not trival (Sutherland et al., 2016). Additionally, anecdotal post-installation reviews1 of these homes have shown that these units are not reaching their full potential because the previous existing heating system continues to operate in parallel with the mini-split. Controllers that communicate between both systems, especially with analytical capabilities, could help this equipment reach its full potential.
5.5 Computation, Controllers and Actuators
There are many ways that computation, controllers and actuators can be improved. One example is variable actuation. Although 2-stage equipment and variable speed fans in air handlers are currently uncommon in residential buildings, these features have the potential to save a lot of energy if used correctly in conjunction with many of the applications discussed in Section 3.0. For example, the DOE test procedure and ASHRAE test procedure in 2006 show that 2-stage furnace control can improve energy efficiency by 3% as compared to single-stage furnaces. One report also showed that even replacement fans with variable speed capacities could save over 150 kWh/year (Aldrich and Williamson, 2014). Being able to control these variable actuators could provide substantial savings opportunities for the U.S.
There are other opportunities besides the central HVAC system where the use of this technology could save energy. For example, the control technologies used in most of the humidity and ventilation (ERV/HRV) systems offer binary control, e.g., on/off, open/closed. Similarly, most of the existing whole- home air purification systems are continuously on, offering no control flexibility. Increasing the control flexibility of these systems could provide more precise control of outputs, reduce oscillations, and increase overall equipment life. Therefore, it is important to develop or integrate technologies that increase the actuation flexibility of all motorized technology, not just the primary air handler.
1Including homes from Efficiency Maine and the Bonneville Power Administration
34
Moreover, increasing control flexibility will also provide an opportunity to correct faults automatically in certain situations, and avoid or delay a technician visit while retaining the system performance.
5.6 Market Barriers A number of market barriers could be addressed to accelerate the uptake of advanced HVAC controls, smart sensors, and their associated firmware/software in the residential market. Consumers consider first cost as the primary obstacle holding them back from purchasing smart HVAC devices that are capable of implementing advanced controls (Danova 2014). While prices for some control devices have declined, prices for high end, effective models are still high (Callaway et al. 2014). The next most common market barrier for consumers is the lack of awareness and education on how devices operate. We have already discussed how this applies to programmable thermostats. Another barrier for consumers is the issue of cyber security and privacy. Consumers might view smart HVAC controls as introducing risk to a device that formerly had none. A report from the California Energy Commission (Ghansah 2012) highlights a few cyber security and privacy issues that may arise from using current programmable communication thermostats and advanced metering infrastructure.
Lastly, for smart HVAC controls to reap additional monetary savings for consumers, they should also have grid interoperability functionality that can take advantage of real-time utility rates or utility incentives. This calls for the proliferation of utility programs and business models that incentivize load- shifting behavior.
35
6.0 References
Aldrich, R and J. Williamson. 2014. “Evaluation of Retrofit Variable-Speed Furnace Fan Motors. DOE-102014-4306. National Renewable Energy Laboratory, Golden, CO. https://www.nrel.gov/docs/fy14osti/60760.pdf
Apex Analytics. 2017. “Energy Trust of Oregon Nest Thermostat Seasonal Savings Pilot Evaluation.” November 22, 2017. Prepared by Apex Analytics, LLC and Demand Side Analytics.
Baechler M and H Hao. 2016. “Business Case for Nonintrusive Load Monitoring.” PNNL-25425, prepared for the Bonneville Power Administration by Pacific Northwest National Laboratory, Richland, Washington.
Brown CA. 2007. “Multizone register controlled residential heating: Optimized for energy use and comfort.” MS Thesis, U. C. Berkeley.
Brown SA, BA Thornton, SH Widder and MC Baechler. 2013. “Review of Residential Low-Load HVAC Systems.” PNNL-23017, Pacific Northwest National Laboratory, Richland, Washington. https://www.pnnl.gov/main/publications/external/technical_reports/PNNL-23017.pdf
Callaway L and N Strother. 2014. “Executive Summary: Smart Thermostats – Communicating Thermostats, Smart Thermostats, and Associated Software and Services: Global Market Analysis and Forecasts.” Navigant Research. Washington, D.C. Accessed on November 12, 2015 at https://www.navigantresearch.com/research/smart-thermostats
Danova T. 2014. The Connected Home Report: Forecasts and Growth Trends FOR THE Leading ‘Internet of Things’ Market. Business Insider Intelligence. Accessed on November 12, 2015 at http://www.businessinsider.com/connected-home-forecasts-and-growth-2014-9
DOE (U.S. Department of Energy). 2016a. “Maintaining Your Air Conditioner.” Accessed on December 30, 2016. https://www.energy.gov/energysaver/maintaining-your-air-conditioner
DOE (U.S. Department of Energy). 2016b. “Overview of Existing and Future Residential Use Cases for Connected Thermostats.” Accessed on December 29, 2016 at https://energy.gov/eere/buildings/downloads/overview-existing-and-future-residential-use-cases- connected-thermostats DOE (U.S. Department of Energy). 2016d. “The National Opportunity for Interoperability and its Benefits for a Reliable, Robust, and Future Grid Realized Through Buildings.” DOE/EE-1341
Domanski PA, HI Henderson, and WV Payne. 2014. “Sensitivity Analysis of Installation Faults on Heat Pump Performance,” NIST Technical Note 1848:9-16.
EERE (Office of Energy Efficiency and Renewable Energy). 2011. Final Rule Technical Support Document: Energy Conservation Standards for Residential Central Air Conditioners, Heat Pumps, and Furnaces. EERE-2011-BT-STD-0011, U.S. Department of Energy, Washington, DC. Accessed on November 12, 2015 at http://www.regulations.gov/#!documentDetail;D=EERE-2011-BT-STD-0011-0012
EIA (Energy Information Administration). 2015. “Residential Energy Consumption Survey (RECS).” U.S. Department of Energy, Washington, D.C.
37
EIA (Energy Information Administration). 2015 “Monthly Energy Review Oct 2015.” U.S. Department of Energy, Washington, D.C. Accessed on November 12, 2015 at http://www.eia.gov/totalenergy/data/monthly/pdf/sec2.pdf
EIA (Energy Information Administration. 2013. How much electricity does an American home use. Accessed November 12, 2015 at http://www.eia.gov/tools/faqs/faq.cfm?id=97&t=3
EIA (Energy Information Administration). 2009. “Residential Energy Consumption Survey (RECS).” U.S. Department of Energy, Washington, D.C.
EPA (U.S. Environmental Protection Agency). 2016. “ENERGY STAR HVAC Quality Installation Program.” Accessed on December 30, 2016. https://www.energystar.gov/ia/home_improvement/downloads/ESQI_factsheet.pdf?8fd5-1967
Gao G and K Whitehouse. 2009. “The Self-Programming Thermostat: Optimizing Setback Schedules based on Home Occupancy Patterns.” In First ACM Workshop on Embedded Sensing Systems for Energy-Efficiency in Buildings, New York, NY.
Ghansah I. 2012. Smart grid cyber security potential threats, vulnerabilities, and risks. California Energy Commission. Accessed on December 22, 2015 at http://www.energy.ca.gov/2012publications/CEC-500-2012-047/CEC-500-2012-047.pdf
Goetzler W, R Zogg, J Young, and J Schmidt. 2012. “Energy Savings Potential and RD&D Opportunities for Residential Building HVAC Systems.” U.S Department of Energy. http://www1.eere.energy.gov/buildings/pdfs/residential_hvac_research_opportunities.pdf
Goyal, S., H. Ingley, and P. Barooah. 2013. “Energy-efficient control of an air handling unit for a single-zone VAV system.” In Proceedings of IEEE 52nd Annual Conference on Decision and Control (CDC), pp. 4796–801. Florence, Italy.
Goyal, S., and P. Barooah. 2015. "Experimental study of occupancy-based control of HVAC zones", Applied Energy, vol. 140, p75-84.
GridTalk. 2013. Shaving Peak Residential Electricity Demand with Grid-Based Technology. Accessed on Nov 12, 2015 at http://www.sandc.com/blogs/index.php/2013/03/shaving-peak-residential- electricity-demand-with-grid-based-technology/
Lstiburek JW and B Pettit. 2010. “Final Report on the Expert Meeting for Diagnostic and Performance Feedback for Residential Space Conditioning System Equipment.” DE-FC26-08NT00601, Building Science Corporation, Westford, Massachusetts.
Maasoumy, M. 2016. “Smart Connected Buildings Design Automation Foundations and Trends.” Foundations and Trends® in Electric Design Automation, Vol 10, No 1-2, pp1-143. http://dx.doi.org/10.1561/1000000043
Oppenheim, P. 1992. “Energy-Saving Potential of a Zoned Forced-Air Heating System.” ASHRAE Transactions.
Parker, D., D. Hoak, et al. 2006. “How Much Energy Are We Using? Potential of Residential Energy Demand Feedback Devices.” 2006 ACEEE Summer Study on Energy Efficiency in Buildings, Asilomar, CA.
38
Piper, B., I. Metzger, N. Ricciardi, and Z. Gao, 2017. “Home Energy Management Pilot System Savings Validation Pilot.” New York State Energy Research and Development Authority. Report Number 17-16. Albany, NY.
Proctor J. 2004. “Residential and Small Commercial Air Conditioning – Rated Efficiency Isn’t Automatic.” In Proceedings of the ASHRAE Winter Meeting, Anaheim, California.
Roth K, D Westphalen, and J Brodrick. 2006. “Residential central AC fault detection and diagnostics.” ASHRAE Journal 48.5, pp. 96–97.
Siemann M. 2010. “Performance of Residential Heating and Cooling Control Strategies using Distributed Wireless Sensor Networks.” MS Thesis. University of Maryland, College Park.
Singer BC, WW Delp, DR Black, and IS Walker. July 2017. “Measured performance of filtration and ventilation systems for fine and ultrafine particles and ozone in an occupied modern California house.” Indoor Air 27.4, pp.780-790.
Somasundaram S, S Katipamula, RG Pratt, ET Mayhorn, BA Akyol, A Somani, N Fernandez, A Steckley, N Foster, and ZT Taylor. 2014. “Transaction-Based Building Controls Framework, Volume 1: Reference Guide.” PNNL-23302, Pacific Northwest National Laboratory, Richland, Washington. http://bgintegration.pnnl.gov/pdf/PNNL_23302.pdf
Sutherland, K, Parker D, Martin, E. 2016. “Evaluation of Mini-Split Heat Pumps as Supplemental and Full System Retrofits in a Hot Humid Climate. FSEC-RR-646-16. Presented at 2016 ACEEE Summer Study on Energy Efficiency in Buildings. http://www.fsec.ucf.edu/en/publications/pdf/fsec-rr-646- 16.pdf
Wang N, and JD Goins. 2015. Buildings of the Future Scoping Study: A Framework for Vision Development. PNNL-24126, Pacific Northwest National Laboratory, Richland, WA.
Walker IS. 2014. “Smart Ventilation (RIVEC) – 2014 BTO Peer Review. Slide 3. https://energy.gov/sites/prod/files/2014/05/f16/rbi09_Singer_042414.pdf
Wiggins, M., and J. Brodrick. 2012. “HVAC fault detection,” ASHRAE Journal, 54(2):78-80.
39